History

liuwei1031 d6cb1a4122 add dynamic C runtime support on windows, test=develop (#19502 )		6 years ago
..
.gitignore	move contrib/inference to paddle/fluid/inference/api	7 years ago
CMakeLists.txt	add dynamic C runtime support on windows, test=develop (#19502 )	6 years ago
README.md	move contrib/inference to paddle/fluid/inference/api	7 years ago
clean.sh	fix manylinux1 Failed to publish artifacts	7 years ago
run.sh	fix spelling errors (#17941 )	6 years ago
simple_on_word2vec.cc	Refine tester of TensorRT engine (#14390 )	6 years ago
trt_mobilenet_demo.cc	AnalysisConfig remove contrib namespace (#15540 )	6 years ago
utils.h	Add INT32 support. INT32 in last switch case	6 years ago
vis_demo.cc	inference_shared_library support profile (#16275 )	6 years ago
windows_inference.md	add guide (#13332 )	6 years ago

Inference Demos

There are several demos:

simple_on_word2vec:
- Follow the C++ codes is in simple_on_word2vec.cc.
- It is suitable for word2vec model.
vis_demo:
- Follow the C++ codes is in vis_demo.cc.
- It is suitable for mobilenet, se_resnext50 and ocr three models.
- Input data format:
  - Each line contains a single record
  - Each record's format is
```
<space splitted floats as data>\t<space splitted ints as shape>
```

To build and execute the demos, simply run

./run.sh $PADDLE_ROOT $TURN_ON_MKL $TEST_GPU_CPU

It will build and execute the demos in both static and shared library.
$PADDLE_ROOT: paddle library path
$TURN_ON_MKL: use MKL or Openblas
$TEST_GPU_CPU: test both GPU/CPU mode or only CPU mode
NOTE: for simple_on_word2vec, must run ctest -R test_word2vec -R to obtain word2vec model at first.